Objective: To examine the association of insulin-like-growth factor 1 (IGF-1) with coronary heart disease (CHD) in women. Methods: Prospective (4 year) case-control study nested within the British Women's Heart and Health Study. With 167 cases and 333 controls. Serum IGF-1 concentrations (on serum stored at -80 degrees C for a median of 4 years) were determined using a two-site immunoenzymometric assay. Results: Among, both cases and controls IGF-1 wits inversely correlated with waist:hip ratio, fasting insulin, HOMA-IR. CRR triglyceride levels and systolic blood pressure, and wits positively correlated with HDL-C levels. The confounder-adjusted (age, socioeconomic position, smoking and physical activity) relative risk ratio for it 1 standard deviation (S.D.) increase in IGF-1 was 0.92 (95% CI: 0.75, 1.12) and additional adjustment for metabolic risk factors (waist:hip ratio, systolic blood pressure, high-density lipoprotein cholesterol. triglycerides, glucose, insulin and C-reactive protein) attenuated this to 0.98 (95% CI: 0.80, 1.21). There was no evidence of non-linear associations and the risk of CHD was similar across quarters of the distribution of IGF-1. Conclusions: Despite associations with established CHD risk factors in this. and other studies, Our findings suggest that higher IGF-1 levels are not associated with CHD in older British women. The contradictory evidence to date in this area highlights the need for further prospective studies and use of genetic epidemiology studies to determine the nature of the association between IGF-1 and CHD. (C) 2008 Elsevier Ireland Ltd. All rights reserved.